Jump to content

Talk:Phrase structure grammar

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

÷ The prior redirection was correct; see http://ieeexplore.ieee.org/iel5/18/22738/01056813.pdf?tp=&isnumber=&arnumber=1056813

See also,

Jurafsky and Martin, "An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition", Prentice Hall Series in Artificial Intelligence, Prentice Hall, Upper Saddle River, NJ.

Page 326-7: "Context-free grammars are also called Phrase Structure Grammars, and the formalism is equivalent to what is also called Backus-Naur form, or BNF."

kraemer 18:20, 18 June 2007 (UTC)[reply]

The redirection was wrong; see [1]

Your references suggest that the redirection (to context-free grammar) should instead be one to formal grammar. A problem with the present article is that it doesn't have much promise to grow as a separate article - at the very least it should be merged with grammar framework (a term I've never heard before). Rp (talk) 15:05, 28 September 2009 (UTC)[reply]
Browsing the Google Books copy confirms this (e.g. p. 421) but I still think they're wrong, or at least, were wrong, to write that phrase structure grammar is a term for context-free grammar. Rp (talk) 23:19, 9 February 2010 (UTC)[reply]

The article says, correctly : "Some authors, however, reserve the term for more restricted grammars in the Chomsky hierarchy: context-sensitive grammars, or context-free grammars". This ambiguity has led to the limitations of context-free grammars being ascribed to the more powerful formal structure at the "top" of the Chomsky hierarchy. I don't whether or how this should be described in Wikipedia. Agingjb (talk) 08:31, 27 July 2010 (UTC)[reply]

See Chomsky 1959 "On certain formal properties of grammar". The phrase structure rewriting rules allow for both context sensitive and context free grammars. On page 139 he says "The intermediate systems are those that assign a phrase structure description to the resulting sentence". This makes it clear the "intermediate systems" (being type 1 and type 2) are the phrase structure grammars. SemMac (talk) 15:19, 11 July 2011 (UTC)[reply]

I think it's even more ovious when consulting "Three models for the description of language" (1956), where "phrase-structure grammars" are grammars with rules Xi -> Yi, where Xi and Yi are strings and Yi must be derived from Xi by replacing a single symbol of Xi with an arbitrary non-empty string. This is similar to the limitations of context-sensitive grammars. (However, a distinct set of terminal symbols is only described as reasonable, not as a requirement, thus the models are not fully compatible). --Zahnradzacken (talk) 09:34, 12 July 2011 (UTC)[reply]

My point was certainly not to say that any ambiguity lies in the work of Chomsky, but that there is a lack of clarity in some interpreters of his work - including the use of the term "phrase structure grammar". Agingjb (talk) 11:34, 13 July 2011 (UTC)[reply]

I made a comment relating to this page (indirectly) here

[edit]

Relating to commonality or differences between phrase structure and dependency grammars. Just a "heads up" :) . — Preceding unsigned comment added by 37.209.42.230 (talk) 14:14, 13 August 20−−−−−−−−−−−−−−15 (UTC)

With the title of the article being "Phrase Structure Grammar," the disproportionate mention of dependency grammar is confusing. I'm sure it was intended to clarify by contrast, but it doesn't. More examples would. Diagram that! Ooze2b (talk) 18:56, 27 March 2016 (UTC)[reply]